repair some problem if gtk-doc fail or such patch for Solaris on new

* doc/Makefile.am: repair some problem if gtk-doc fail or such
* configure.in: patch for Solaris on new autoconf closes #98880
* doc/parsedecl.py: repair the frigging API building script,
  did I say that python xmllib sucks ?
* doc/libxml2-api.xml doc/libxml2-refs.xml: regenerated, reordering
  and some comment are no more truncated.
Daniel

ChangeLog

@@ -1,3 +1,12 @@
+Tue Nov 19 23:25:47 CET 2002 Daniel Veillard <daniel@veillard.com>
+
+	* doc/Makefile.am: repair some problem if gtk-doc fail or such
+	* configure.in: patch for Solaris on new autoconf closes #98880 
+	* doc/parsedecl.py: repair the frigging API building script,
+	  did I say that python xmllib sucks ?
+	* doc/libxml2-api.xml doc/libxml2-refs.xml: regenerated, reordering
+	  and some comment are no more truncated.
+
 Tue Nov 19 09:09:04 CET 2002 Daniel Veillard <daniel@veillard.com>
 
 	* parser.c: Martin Stoilov pointed out a potential leak in

configure.in

@@ -82,12 +82,63 @@ LDFLAGS=${_ldflags}
 dnl Checks for header files.
 AC_HEADER_DIRENT
 AC_HEADER_STDC
-AC_CHECK_HEADERS(fcntl.h unistd.h ctype.h dirent.h errno.h malloc.h)
+AC_CHECK_HEADERS([fcntl.h])
-AC_CHECK_HEADERS(stdarg.h sys/stat.h sys/types.h time.h ansidecl.h)
+AC_CHECK_HEADERS([unistd.h])
-AC_CHECK_HEADERS(ieeefp.h nan.h math.h fp_class.h float.h)
+AC_CHECK_HEADERS([ctype.h])
-AC_CHECK_HEADERS(stdlib.h sys/socket.h netinet/in.h arpa/inet.h)
+AC_CHECK_HEADERS([dirent.h])
-AC_CHECK_HEADERS(netdb.h sys/time.h sys/select.h sys/mman.h)
+AC_CHECK_HEADERS([errno.h])
-AC_CHECK_HEADERS(sys/timeb.h signal.h arpa/nameser.h resolv.h)
+AC_CHECK_HEADERS([malloc.h])
+AC_CHECK_HEADERS([stdarg.h])
+AC_CHECK_HEADERS([sys/stat.h])
+AC_CHECK_HEADERS([sys/types.h])
+AC_CHECK_HEADERS([time.h])
+AC_CHECK_HEADERS([ansidecl.h])
+AC_CHECK_HEADERS([ieeefp.h])
+AC_CHECK_HEADERS([nan.h])
+AC_CHECK_HEADERS([math.h])
+AC_CHECK_HEADERS([fp_class.h])
+AC_CHECK_HEADERS([float.h])
+AC_CHECK_HEADERS([stdlib.h])
+AC_CHECK_HEADERS([sys/socket.h], [], [],
+[#if HAVE_SYS_TYPES_H
+# include <sys/types.h>
+# endif
+])
+AC_CHECK_HEADERS([netinet/in.h], [], [],
+[#if HAVE_SYS_TYPES_H
+# include <sys/types.h>
+# endif
+])
+AC_CHECK_HEADERS([arpa/inet.h], [], [],
+[#if HAVE_SYS_TYPES_H
+# include <sys/types.h>
+# endif
+#if HAVE_ARPA_INET_H
+# include <arpa/inet.h>
+# endif
+])
+AC_CHECK_HEADERS([netdb.h])
+AC_CHECK_HEADERS([sys/time.h])
+AC_CHECK_HEADERS([sys/select.h])
+AC_CHECK_HEADERS([sys/mman.h])
+AC_CHECK_HEADERS([sys/timeb.h])
+AC_CHECK_HEADERS([signal.h])
+AC_CHECK_HEADERS([arpa/nameser.h], [], [],
+[#if HAVE_SYS_TYPES_H
+# include <sys/types.h>
+# endif
+])
+AC_CHECK_HEADERS([resolv.h], [], [],
+[#if HAVE_SYS_TYPES_H
+# include <sys/types.h>
+# endif
+#if HAVE_NETINET_IN_H
+# include <netinet/in.h>
+# endif
+#if HAVE_ARPA_NAMESER_H
+# include <arpa/nameser.h>
+# endif
+])
 
 dnl Specific dir for HTML output ?
 if test "x$with_html_dir" = "x" ; then

doc/Makefile.am

@@ -37,21 +37,19 @@ $(APIPAGES): libxml2-refs.xml site.xsl api.xsl
 	  $(bindir)/xsltproc --html $(top_srcdir)/doc/api.xsl $(top_srcdir)/doc/xml.html ; fi );
 
 scan:
-	gtkdoc-scan --module=libxml --source-dir=$(DOC_SOURCE_DIR) --ignore-headers="acconfig.h config.h win32config.h trio.h triostr.h triop.h config-mac.h XMLTestPrefix2.h  XMLTestPrefix.h triodef.h trionan.h xlink.h libxml.h libxml2-py.h libxml_wrap.h xmlschemas.h xmlschemastypes.h"
+	-gtkdoc-scan --module=libxml --source-dir=$(DOC_SOURCE_DIR) --ignore-headers="acconfig.h config.h win32config.h trio.h triostr.h triop.h config-mac.h XMLTestPrefix2.h  XMLTestPrefix.h triodef.h trionan.h xlink.h libxml.h libxml2-py.h libxml_wrap.h xmlschemas.h xmlschemastypes.h"
 
 templates: scan
-	gtkdoc-mktmpl --module=libxml
+	-gtkdoc-mktmpl --module=libxml
 
-sgml:
+sgml: templates
-	gtkdoc-mkdb --module=libxml --source-dir=$(DOC_SOURCE_DIR)
+	-gtkdoc-mkdb --module=libxml --source-dir=$(DOC_SOURCE_DIR)
 
-xml: sgml
+xml: sgml 
 	if test -d xml ; then rm xml/* ; else mkdir xml ; fi
 	-@(for s in sgml/*.sgml ; do name=`basename $$s .sgml` ; \
-	   echo '<!DOCTYPE refentry [ <!ENTITY nbsp   "&#x00A0;"> ]>' > \
+	   xmllint --sgml $$s > xml/"$$name".xml ; done)
-	       xml/"$$name".xml ;\
+
-	   grep -v "^<colspec" $$s | sed 's+<anchor(.*>+<anchor\1/>+' >> xml/"$$name".xml ; done)
-	   
 libxml2-api.xml libxml2-refs.xml: xml parsedecl.py libxml-decl.txt libxml-decl-list.txt
 	-(./parsedecl.py)
 
@@ -67,7 +65,7 @@ clean-local:
 maintainer-clean-local: clean
 	rm -rf sgml html libxml-decl-list.txt libxml-decl.txt 
 
-libxml-decl-list.txt : templates
+libxml-decl-list.txt libxml-decl.txt : templates
 
 libxml-sections.txt : scan
 	cp libxml-decl-list.txt libxml-sections.txt

doc/parsedecl.py

@@ -450,196 +450,102 @@ def insertComment(name, title, value, id):
     nbcomments = nbcomments + 1
 
 
+import libxml2
 import os
-import xmllib
-try:
-    import sgmlop
-except ImportError:
-    sgmlop = None # accelerator not available
-
-debug = 0
-
-if sgmlop:
-    class FastParser:
-	"""sgmlop based XML parser.  this is typically 15x faster
-	   than SlowParser..."""
-
-	def __init__(self, target):
-
-	    # setup callbacks
-	    self.finish_starttag = target.start
-	    self.finish_endtag = target.end
-	    self.handle_data = target.data
-
-	    # activate parser
-	    self.parser = sgmlop.XMLParser()
-	    self.parser.register(self)
-	    self.feed = self.parser.feed
-	    self.entity = {
-		"amp": "&", "gt": ">", "lt": "<",
-		"apos": "'", "quot": '"'
-		}
-
-	def close(self):
-	    try:
-		self.parser.close()
-	    finally:
-		self.parser = self.feed = None # nuke circular reference
-
-	def handle_entityref(self, entity):
-	    # <string> entity
-	    try:
-		self.handle_data(self.entity[entity])
-	    except KeyError:
-		self.handle_data("&%s;" % entity)
-
-else:
-    FastParser = None
 
 
-class SlowParser(xmllib.XMLParser):
+def analyzeXMLDescriptionRow(doc, desc, id, row):
-    """slow but safe standard parser, based on the XML parser in
+    if doc == None or desc == None or id == None or row == None:
-       Python's standard library."""
+        return
+    ctxt = doc.xpathNewContext()
+    ctxt.setContextNode(row)
+    param = ctxt.xpathEval("entry[1]/parameter")
+    entries = ctxt.xpathEval("entry")
+    if param == []:
+        is_param = 0
+	name = None
+    else:
+        name = param[0].content
+	is_param = 1
+    str = entries[1].content
+    str = string.replace(str, '\n', ' ')
+    str = string.replace(str, '\r', ' ')
+    str = string.replace(str, '    ', ' ')
+    str = string.replace(str, '   ', ' ')
+    str = string.replace(str, '  ', ' ')
+    while len(str) >= 1 and str[0] == ' ':
+	str=str[1:]
 
-    def __init__(self, target):
+    insertParameterComment(id, name, str, is_param)
-	self.unknown_starttag = target.start
-	self.handle_data = target.data
-	self.unknown_endtag = target.end
-	xmllib.XMLParser.__init__(self)
 
-def getparser(target = None):
-    # get the fastest available parser, and attach it to an
-    # unmarshalling object.  return both objects.
-    if target == None:
-	target = docParser()
-    if FastParser:
-	return FastParser(target), target
-    return SlowParser(target), target
 
-class docParser:
-    def __init__(self):
-        self._methodname = None
-	self._data = []
-	self.id = None
-	self.title = None
-	self.descr = None
-	self.string = None
 
-    def close(self):
+def analyzeXMLDescription(doc, desc):
-        if debug:
+    if doc == None or desc == None:
-	    print "close"
+        return
+    ctxt = doc.xpathNewContext()
+    ctxt.setContextNode(desc)
 
-    def getmethodname(self):
+    #
-        return self._methodname
+    # get the function name
+    #
+    try:
+	title = ctxt.xpathEval("title")[0].content
+    except:
+        return
+    old_id = ctxt.xpathEval("string(title/anchor/@id)")
+    id = string.replace(title, '(', ' ');
+    id = string.replace(id, ')', ' ');
+    id = string.split(id) [0]
 
-    def data(self, text):
+    #
-        if debug:
+    # get the function comments
-	    print "data %s" % text
+    #
-        self._data.append(text)
+    comment = ""
+    paras = ctxt.xpathEval("para")
+    for para in paras:
+        str = para.content
+	str = string.replace(str, '\n', ' ')
+	str = string.replace(str, '\r', ' ')
+	str = string.replace(str, '    ', ' ')
+	str = string.replace(str, '   ', ' ')
+	str = string.replace(str, '  ', ' ')
+	while len(str) >= 1 and str[0] == ' ':
+	    str=str[1:]
 
-    def start(self, tag, attrs):
+        comment = comment + str
-        if debug:
-	    print "start %s, %s" % (tag, attrs)
-	if tag == 'refsect2':
-	    self.id = None
-	    self.title = None
-	    self.descr = None
-	    self.string = None
-	    self.type = None
-	    self.in_parameter = 0
-	    self.is_parameter = 0
-	    self.parameter = None
-	    self.parameter_info = None
-	    self.entry = 0
-	elif tag == 'para':
-	    self._data = []
-	elif tag == 'title':
-	    self._data = []
-	elif tag == 'tgroup':
-	    self.in_parameter = 1
-	elif tag == 'row':
-	    self._data = []
-	    self.entry = 0
-	elif tag == 'entry':
-	    self.entry = self.entry + 1
-	elif tag == 'parameter' and self.in_parameter == 1:
-	    self._data = []
-	elif tag == 'anchor' and self.id == None:
-	    if attrs.has_key('id'):
-	        self.orig_id = attrs['id']
-		self.id = string.replace(self.orig_id, '-CAPS', '')
-		self.id = string.replace(self.id, '-', '_')
 
-    def end(self, tag):
+    insertComment(id, title, comment, old_id)
-        if debug:
+
-	    print "end %s" % tag
+    rows = ctxt.xpathEval("informaltable/tgroup/tbody/row")
-	if tag == 'refsect2':
+    for row in rows:
-	    self.type = insertComment(self.id, self.title, self.string,
+        analyzeXMLDescriptionRow(doc, desc, id, row)
-	                              self.orig_id)
+
-	    self.string = None
+def analyzeXMLDoc(doc):
-	elif tag == 'row':
+    if doc == None:
-	    if self.parameter_info != None and self.parameter_info != '':
+        return
-		insertParameterComment(self.id, self.parameter,
+    ctxt = doc.xpathNewContext()
-				       self.parameter_info, self.is_parameter)
+    descriptions = ctxt.xpathEval("//refsect2")
-	    self.parameter_info = None
+    print len(descriptions)
-	    self.parameter = 0
+    for description in descriptions:
-	    self.is_parameter = 0
+        analyzeXMLDescription(doc, description)
-	elif tag == 'parameter' and self.in_parameter == 1 and self.entry == 1:
+
-	    str = ''
+doc = libxml2.parseFile("xml/c14n.xml");
-	    for c in self._data:
+analyzeXMLDoc(doc)
-		str = str + c
+doc.freeDoc()
-	    str = string.replace(str, '\n', ' ')
-	    str = string.replace(str, '\r', ' ')
-	    str = string.replace(str, '    ', ' ')
-	    str = string.replace(str, '   ', ' ')
-	    str = string.replace(str, '  ', ' ')
-	    while len(str) >= 1 and str[0] == ' ':
-		str=str[1:]
-	    self.parameter = str
-	    self.is_parameter = 1
-	    self._data = []
-	elif tag == 'para' or tag == 'entry':
-	    str = ''
-	    for c in self._data:
-		str = str + c
-	    str = string.replace(str, '\n', ' ')
-	    str = string.replace(str, '\r', ' ')
-	    str = string.replace(str, '    ', ' ')
-	    str = string.replace(str, '   ', ' ')
-	    str = string.replace(str, '  ', ' ')
-	    while len(str) >= 1 and str[0] == ' ':
-		str=str[1:]
-	    if self.string == None:
-		self.string = str
-	    elif self.in_parameter == 1:
-		self.parameter_info = str
-	    self._data = []
-	elif tag == 'title':
-	    str = ''
-	    for c in self._data:
-	        str = str + c
-	    str = string.replace(str, '\n', ' ')
-	    str = string.replace(str, '\r', ' ')
-	    str = string.replace(str, '    ', ' ')
-	    str = string.replace(str, '   ', ' ')
-	    str = string.replace(str, '  ', ' ')
-	    while len(str) >= 1 and str[0] == ' ':
-		str=str[1:]
-	    self.title = str
 
 xmlfiles = 0
 filenames = os.listdir("xml")
 for filename in filenames:
+    print filename
     try:
-        f = open("xml/" + filename, 'r')
+        doc = libxml2.parseFile("xml/" + filename)
-    except IOError, msg:
+	analyzeXMLDoc(doc)
-        print file, ":", msg
+	doc.freeDoc()
+	xmlfiles = xmlfiles + 1
+    except:
+        print "failed to parse XML description %s" % ("xml/" + filename)
 	continue
-    data = f.read()
-    (parser, target)  = getparser()
-    parser.feed(data)
-    parser.close()
-    xmlfiles = xmlfiles + 1
 
 print "Parsed: %d XML files collexting %d comments" % (xmlfiles, nbcomments)